Pipeline for the Zebrafish Model as a Predictor of Specific Teratogenic Effects

Abstract

Over 80,000 synthetic chemicals are currently in commercial use without a full assessment of their adverse effects, especially on developmental stages of life. Exposure to some of these chemicals has been linked to birth defects, chronic disease and cancer. Synthetic chemicals are also present in everyday items, such as food, household cleaners and lawn care products. Zebrafish (Danio rerio) are an effective alternative in vivo model to standard mammalian models for teratogenic studies owing to their rapid development, low cost, and genetic homology to humans. The availability of transgenic fish that express fluorescence in specific tissues or organs allows for detection of tissue-specific structural malformations in live fish. We have used the zebrafish model to understand the adverse effects that potential teratogens have on living systems, and to establish a rapid and reliable screening method in vivo. Transgenic fish were used to screen for chemicals that perturb angiogenesis and posterior lateral line (PLL) development, and the screening results subjected to computational toxicology to elucidate the molecular pathways by which teratogens cause these specific phenotypic effects. We screened 175 ToxCast Phase I chemicals in zebrafish for vascular teratogenicity (vascular disrupting chemicals; VDCs) and identified 10 chemicals that interfered with intersegmental vessel (ISV) development. For PLL teratogenicity (PLL disruptors, PLLDs), we tested all 292 chemicals from the ToxCast Phase I library and identified 22 chemicals that reduced the number of deposited neuromasts. The chemical hit lists for each of the developmental perturbations observed were then compared to the ToxCast in vitro assays to produce a phenotype-specific assay profile of genes or molecular pathways through which these chemicals may be acting. The assay profiles were used to predict other chemicals from the ToxCast Phase II library that could be potential VDCs or PLLDs. In summary, we have established a rapid screening method for phenotypic outcomes in zebrafish and combined it with computational analysis of ToxCast in vitro data to obtain phenotype-specific assay profiles that can further be used to rank toxicants tested in the Tox21 project to speed up prioritization of chemicals for further testing in vivo, and in the long run result in an improved risk assessment for human and environmental health.

Description

Keywords

Posterior lateral line, Neuromast disruptor, Zebrafish, Transgenic, Serotonin, High throughput screening, Computational toxicology, Pipeline, Teratogens, Toxicity prediction, Vascular disruptor, Vasculature

Citation